- Email: [email protected]
Perfectionism and self-consciousness in social phobia and panic disorder with agoraphobia
BEHAVIOUR RESEARCH AND THERAPY
PERGAMON
Behaviour Research and Therapy 37 (1999) 799±808
www.elsevier.com/locate/brat
Perfectionism and self-consciousness in social phobia and panic disorder with agoraphobia Fredrik Saboonchi *, Lars-Gunnar Lundh, Lars-GoÈran OÈst Department of Psychology, Stockholm University, 106 91 Stockholm, Sweden Accepted 22 September 1998
Abstract Social phobics were compared to patients with panic disorder with agoraphobia and normal controls on perfectionism and self-consciousness. On concern over mistakes and doubts about action, social phobics scored higher than patients with panic disorder. Social phobics also demonstrated a higher level of public self-consciousness than patients with panic disorder and when this dierence was controlled for the signi®cant dierences on perfectionism disappeared. Within each patient group, however, perfectionism was more robustly related to social anxiety than was public self-consciousness, which replicates the ®ndings of Saboonchi and Lundh [Saboonchi, F. & Lundh, L. G. (1997). Perfectionism, self-consciousness and anxiety. Personality and Individual Dierences, 22, 921±928.] from a non-clinical sample. The results are discussed in terms of public self-consciousness being a dierentiating characteristic of the more severe kind of social anxiety which is typical of social phobia. # 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction Research on perfectionism has established a wide array of associations between various aspects of perfectionism and numerous forms of psychopathology, including depression (Hewitt & Flett, 1991a, 1993; Blatt, Quinlan, Pilkonis & Shea, 1995), social phobia (Juster et al., 1996; Lundh & OÈst, 1996), obsessive±compulsive disorder (Frost & Steketee, 1997), eating disorders (e.g. Joiner, Heatherton, Rudd & Schmidt, 1997) and borderline personality disorder (Hewitt, Flett & Turnbull, 1994). There are at present two main multidimensional approaches to the empirical study of perfectionism. In the Frost, Marten, Lahart, and Rosenblate (1990) approach, perfectionism is * Corresponding author. E-mail: [email protected] 0005-7967/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 5 - 7 9 6 7 ( 9 8 ) 0 0 1 8 3 - 1
800
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
de®ned as consisting of dimensions of personal standards, concern over mistakes, doubts about the quality of one's own actions, organisation, and parental expectations and criticism. Of these dimensions, concern over mistakes (CM), doubts about action (DA) and parental criticism (PC) have been found to be more associated with psychopathology. In the Hewitt and Flett (1991a) approach, on the other hand, three dimensions of perfectionism are identi®ed: self-oriented perfectionism, other-oriented perfectionism and socially prescribed perfectionism. The latter two dimensions address the interpersonal aspects of perfectionistic demands and socially prescribed perfectionism has been found to be more likely to be related to psychopathology. It can be asked whether perfectionism is a general characteristic of psychopathology or whether dierent forms of psychopathology dier on perfectionism or on some dimensions of perfectionism. With regard to anxiety disorders, higher levels of concern over mistakes, doubts about action and parental criticism were reported in social phobics compared to controls by Juster et al. (1996) and by Lundh and OÈst (1996). Frost and Steketee (1997) found that both patients with obsessive±compulsive disorder (OCD) and patients with panic disorder with agoraphobia had higher levels of concern over mistakes than the controls, but that they did not dier from one another. On doubts about action, the pattern was dierent; OCD patients scored higher than both panic patients and normal controls, who did not dier from one another. On parental criticism, ®nally, the panic patients scored higher than both the OCD patients and the controls, who did not dier from one another. The main purpose of the present study was to compare patients with social phobia, panic disorder with agoraphobia and normal controls on the Frost et al. (1990) Multidimensional Perfectionism Scale. Since social anxiety by de®nition involves a preoccupation with one's social presentation and doubts about one's ability to convey a desired impression on other persons (Schlenker & Leary, 1982; Clark & Wells, 1995; Rapee & Heimberg, 1997), a certain overlap with perfectionistic concerns is apparent. According to the Rapee and Heimberg (1997, pp. 743±744) conceptualization of social phobia, for example, the social phobic ``formulates a prediction of the performance standard or norm which he/she expects the audience to utilize in the given situation. The discrepancy between the person's perception of the audience's appraisal of his/ her performance (appearance and/or behavior) and the person's perception of the audience's standard for the evaluation of his/her appearance and/or behavior, determines the perceived likelihood of negative evaluation from the audience''. The more perfectionistic standards the individual has, or believes that the audience has, the more likely will a negative evaluation appear, which is what the social phobic fears. With these clear indications of associations between social anxiety and perfectionism, two questions can be raised. First, is perfectionism more speci®cally related to social phobia than other psychopathological forms of anxiety? Second, what aspects of perfectionism account for this relationship? Although in a non-clinical population, a similar issue was attended to earlier by Saboonchi and Lundh (1997). Several dimensions of perfectionism, primarily concern over mistakes (CM), doubts about actions (DA) and socially prescribed perfectionism (SPP), showed strong patterns of relationship with a variety of anxiety measures (social anxiety, agoraphobic fears, and fears of bodily injury, death and illness). Social anxiety, however, correlated stronger than agoraphobic fears with CM and stronger than fears of bodily injury,
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
801
death and illness with CM and socially prescribed perfectionism. The present study aims to test the hypothesis that social phobia involves more of perfectionism than panic disorder with agoraphobia, by comparing two groups of patients with these diagnoses and a control group on the Frost et al. (1990) Multidimensional Perfectionism Scale. Another concept that is relevant in this context is that of public self-consciousness. Public self-consciousness is an aspect of self-awareness de®ned by Fenigstein, Scheier, and Buss (1975) as re¯ecting awareness of the self as a social object. This de®nition addresses the preoccupation with, and the degree of attention directed to, aspects of the self that are accessible to others such as looks, impressions etc. Social anxiety has been found to correlate positively with public self-consciousness (Fenigstein et al., 1975; Hope & Heimberg, 1988) and individuals with social phobia have been found to score higher on public self-consciousness than normal controls and individuals with other anxiety disorders (Bruch, Heimberg, Berger & Collins, 1989; Bruch & Heimberg, 1994). In the earlier study by Saboonchi and Lundh (1997), however, ®ndings indicated a non-signi®cant pattern of relationship between public self-consciousness and social anxiety when dimensions of perfectionism were controlled for. Since these results were found in a non-clinical sample of subjects, it was unclear whether the same results would be obtained with a clinical sample of patients with a diagnosis of social phobia. A second purpose of the present paper, therefore, was to explore this issue. In summary, three hypotheses are tested by this study: (a) both social phobia and panic disorder with agoraphobia are characterized by elevated degrees of perfectionism, particularly on the subscales concern over mistakes, doubts about action and parental criticism. (b) Social phobia is characterized by a higher degree of perfectionism than panic disorder with agoraphobia. (c) Social phobia is also characterized by higher scores on public selfconsciousness than both panic disorder with agoraphobia and normal controls; this dierence, however, disappears when degree of perfectionism is controlled for. On the other hand, the dierences on perfectionism remain when public self-consciousness is controlled for.
2. Method 2.1. Subjects The subjects were 52 social phobics (8 males and 44 females aged 33.6 (S.D. = 8.8) yr), 55 patients with a diagnosis of panic disorder with agoraphobia (19 males and 36 females with a mean age of 34.5 (S.D. = 10.7) yr) and 113 control subjects (27 males and 86 females aged 30.6 (S.D. = 9.8) yr). The social phobics were part of a treatment study with cognitivebehaviour therapy, and as part of that study they were interviewed by trained clinical psychologists with the revised version of the Anxiety Disorders Interview Schedule (ADIS-R; DiNardo & Barlow, 1988); all patients met the DSM-III-R criteria (American Psychiatric Association, 1987) for social phobia. The panic patients were part of another treatment study with cognitive-behaviour therapy and as part of that study they were interviewed by a trained clinical psychologist with the Anxiety Disorders Interview Schedule (ADIS-IV; Brown, DiNardo & Barlow, 1994); all of the patients met the DMS-IV criteria (American Psychiatric Association, 1994) for panic disorder with agoraphobia. Potential Ss in both studies were
802
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
excluded if they received a diagnosis of primary depression, bipolar disorder, psychotic disorder or active drug or alcohol dependence. The control group were volunteers, who were recruited in two ways: (a) with advertisements in a living area adjacent to the university (36 females and 7 males); these subjects were paid for their participation and (b) undergraduate students in psychology (50 females and 20 males), who received course credit for their participation. All patients were administered the questionnaires before they started therapy. 2.2. Psychometric measures 2.2.1. Multidimensional perfectionism scale (MPS) The MPS (Frost et al., 1990) contains 35 items in the form of statements with a Likert type 5-point response format from `strongly disagree' to `strongly agree'. These 35 items are being divided into 6 subscales tapping six dimensions of perfectionism, i.e. concern over mistakes, personal standards, parental expectations, parental criticism, doubts about actions and organisation. 2.2.2. Self-consciousness scale (SCS) This instrument was a Swedish translation (Nystedt & Smari, 1989) of the Fenigstein et al. (1975) Self-Consciousness Scale, which includes a subscale for private self-consciousness (i.e. the disposition to attend to one's own thoughts, feelings, moods and attitudes), a subscale for public self-consciousness (i.e. the disposition to focus on outwardly observable aspects of the self, such as physical appearance and overt behaviour) and a subscale of social anxiety. 2.2.3. State-Trait Anxiety Inventory, Trait Form (STAI-T) The STAI-T (Spielberger, Gorsuch & Lushene, 1970) contains 20 items that assess enduring symptoms of anxiety. 2.2.4. Beck Depression Inventory (BDI) The BDI (Beck, Ward, Mendelson, Mock & Erbaugh, 1961) is a 21-item self-report questionnaire that assesses the presence and severity of cognitive, behavioural, aective and somatic symptoms of depression. Each item consists of a 0 to 3 self-rating, with increasing scores indicating greater symptom severity.
3. Results Means and standard deviations for the two groups of patients and controls are presented in Table 1. As can be seen, there were no scores available on BDI or STAI-T for the control group; accordingly, the analysis involving these measures was only performed on the two patient groups. A MANOVA on all the subscales of the MPS and SCS with all the three groups revealed a highly signi®cant eect of Group (Wilks' Lambda = 0.613, F(18, 416) = 6.41, P < 0.001). As is displayed in Table 1, the follow-up univariate tests revealed signi®cant eects of group on
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
803
Table 1 Means and standard deviations for social phobics, panic patients and control subjects Social phobics
Panic patients
Controls
M
S.D.
M
S.D.
M
S.D.
F-value
100.7 26.7a 10.0a 21.8 9.6a 11.8 20.8
21.6 7.6 3.5 6.0 4.6 5.4 4.5
91.0 22.7b 8.4b 19.6 9.3a 11.4 19.6
24.3 7.8 3.3 6.2 4.8 5.8 5.1
84.4 19.7c 7.5b 20.5 7.0b 10.1 19.5
17.9 6.4 3.2 5 3.6 4.3 5.2
17.43* 10.10* 2.11 9.36* 2.86 2.33
SCS Public self-consciousness Private self-consciousness Social anxiety
22.8a 25.8 18.1a
3.9 6.4 4.4
18.2b 23.9 12.3b
5.2 5.4 5.7
17.8b 23.6 9.8c
5.1 6.7 5.5
19.80* 4.57 43.41*
Beck Depression Inventory State-Trait Anxiety Inventory-Trait
16.8 53.3
9.2 8.7
13.4 49.7
8.3 10.2
MPS Total Concern over mistakes Doubts about actions Personal standards Parental criticism Parental expectation Organization
± ±
ANOVA
± ±
3.79 3.83
*
P < 0.001.Means with dierent superscripts are signi®cantly dierent at the 0.05 level.
social anxiety, public self-consciousness, concern over mistakes, doubts about action and parental criticism. To examine these dierences further, one-way ANOVAS were performed on these MPS and SCS subscales and multiple comparisons among the three scores were done using Schee Ftest. The results showed that the social phobics scored signi®cantly higher than the controls on social anxiety, public self-consciousness, concern over mistakes, doubts about action and parental criticism. In addition, the social phobics scored signi®cantly higher than the panic patients on social anxiety, public self-consciousness, concern over mistakes and doubts about action. The panic patients, in turn, scored signi®cantly higher than the controls on parental criticism, social anxiety and concern over mistakes. The two patient groups were compared on the Beck Depression Inventory and the StateTrait Anxiety Inventory-Trait version by means of a two-tailed t-test. The social phobics had a non-signi®cant tendency to higher scores on both the BDI, t(104) = 1.95, P = 0.054, and the STAI-T, t(104) = 1.96, P = 0.053. In order to test the hypothesis that the group dierences on perfectionism between the two patient groups would remain when public self-consciousness was controlled for, a multivariate analysis of covariance on concern over mistakes and doubts about action was carried out (since these were the only dimensions of perfectionism on which social phobics scored higher than panic patients), with public self-consciousness included as a covariate. The multivariate test and both univariate tests were non-signi®cant, i.e. no signi®cant dierences remained, neither on concern over mistakes, F(1, 103) = 0.08, nor on doubts about action, F(1, 103) = 0.17. However, when analysis of covariance was performed on public self-consciousness
804
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
Table 2 Correlations and partial correlations (controlled for public self-consciousness) between aspects of perfectionism and measures of anxiety and depression Social phobics
SCS social anxiety STAI-T BDI
Panic patients
Controls
CM
DA
CM
DA
CM
DA
0.32* (0.23) 0.33* (0.28*) 0.30* (0.28*)
0.44** (0.36*) 0.53** (0.50**) 0.18 (0.15)
0.46* (0.21) 0.23 (0.26) 0.27 (0.12)
0.34** (0.21) 0.29* (0.26) 0.21 (0.12)
0.48** (0.43**) ± ±
0.54** (0.50**) ± ±
Values in parentheses are partial correlations controlling for SCS public self-consciousness. CM = concern over mistakes, DA = doubts about action.*P < 0.05.**P < 0.01.
with concern over mistakes and doubts about action as covariates, the group eect remained highly signi®cant, F(1, 102) = 16.24, P < 0.001. Regarding the comparisons with the controls, however, when a similar MANOVA and its follow up univariate tests on the subscales of MPS were performed with public selfconsciousness included as covariate, the dierences that were found in the initial analysis between each patient group and the controls remained signi®cant. Correlations and partial correlations (controlling for public self-consciousness) between CM, DA and measures of anxiety (social anxiety and STAIT-T) and depression (BDI) for the two patient groups and the controls (only social anxiety) are presented in Table 2. For social phobics CM correlated signi®cantly with both measures of anxiety and with depression (P < 0.05). With the exception for social anxiety these correlations remained signi®cant when public self-consciousness was controlled for. DA correlated signi®cantly with social anxiety and STAIT-T and these correlations remained signi®cant (P < 0.01) when controlled for public self-consciousness. The panic patients demonstrated a somewhat dierent pattern. CM correlated signi®cantly (P < 0.01) with SCS social anxiety and this correlation remained when controlled for public self-consciousness. DA correlated signi®cantly (P < 0.05) with SCS social anxiety and STAITT. These correlations, however, were rendered non-signi®cant when public self-consciousness was controlled for. Concerning the controls, both CM and DA correlated signi®cantly with SCS social anxiety (P < 0.01) and these correlations still remained signi®cant (P < 0.01) when public selfconsciousness was controlled for. Correlations and partial correlations (controlling for concern over mistakes and doubts about action) between public self-consciousness and measures of anxiety and depression are displayed in Table 3. For social phobics public self-consciousness correlated signi®cantly with social anxiety; when CM and DA were controlled for, however, this correlation became non-signi®cant. For panic patients there were signi®cant correlations between public self-consciousness and social anxiety (P < 0.01) and BDI (P < 0.05). The latter, however, disappeared when CM and DA was controlled for. For the controls public self-consciousness correlated signi®cantly with SCS
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
805
Table 3 Correlations and partial correlations (controlled for concern over mistakes and doubts about action) between public self-consciousness and measures of anxiety and depression Public self-consciousness
SCS social anxiety State-Trait Anxiety Inventory-Trait Beck Depression Inventory
panic patients
social phobics
controls
0.39** (0.18) 0.18 (0.05) 0.29* (0.19)
0.33** (0.16) 0.20 (ÿ0.03) 0.11 (ÿ0.02)
0.28** (0.11) ± ±
Values in parentheses are partial correlations controlling for concern over mistakes and doubt about actions subscales of MPS.*P < 0.05.**P < 0.01.
social anxiety ( p < 0.01). This correlation, however, was no longer signi®cant when CM and DA were controlled for.
4. Discussion The results of the present study showed that the social phobics scored higher than the normal controls on three dimensions of perfectionism, i.e. concern over mistakes, doubts about action and parental criticism. They also scored higher than patients with a diagnosis of panic disorder with agoraphobia on two of these dimensions, i.e. concern over mistakes and doubts about action. The panic patients, however, also showed evidence of an elevated degree of perfectionism, since they scored higher than the controls on concern over mistakes and parental criticism. These results are consistent with earlier studies, both with regard to social phobics (e.g. Juster et al., 1996) and panic patients (Frost & Steketee, 1997). The results also supported the hypothesis that social phobia involves more of perfectionism than panic disorder, at least with regard to the dimensions concern over mistakes and doubts about action. In addition, the social phobics scored higher than both the controls and the panic patients on social anxiety and on public self-consciousness. It is interesting to note that, although the panic patients scored higher than the controls on social anxiety, they did not dier from the controls on the public self-consciousness scale. This indicates that the degree of public selfconsciousness, i.e. the individual's tendency to focus on how he/she appears to others, is a highly dierentiating characteristic between social phobics and panic patients. The two patient groups, in fact, diered considerably more on public self-consciousness than on the dimensions of perfectionism and the hypothesis that the group dierences between the two patient groups on perfectionism would remain when public self-consciousness was controlled for was ¯atly contradicted. Quite to the contrary, all the group dierences on perfectionism disappeared when the eects of public self-consciousness were controlled. Moreover, the social phobics' higher scores on public self-consciousness, as compared with the panic patients, remained when perfectionism was controlled for. This suggests that the social phobics' stronger tendency to focus on how they appear to others, as compared with the
806
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
panic patients', is not reducible to their tendency to doubt the quality of their behaviour and the harshness that characterizes their perfectionistic concern over mistakes. The hypothesis that this dierence is determined by the perfectionistic aspects of this trait got no support whatever. On the contrary, the results suggest that social phobia is essentially characterized by a higher level of public self-consciousness than panic disorder with agoraphobia, in a way which is not motivated by perfectionistic concerns. This result, at ®rst sight, seems to be totally at odds with the results in the study by Saboonchi and Lundh (1997), where the correlation between social anxiety and perfectionism remained when public self-consciousness was controlled for, whereas the correlation between social anxiety and public self-consciousness was no longer signi®cant when perfectionism was controlled for. One dierence between that study and the present one is that Saboonchi and Lundh (1997) used a non-clinical sample. This might suggest that the relationships between social anxiety, perfectionism and public self-consciousness are dierent in clinical and nonclinical samples. What speaks against this conclusion, however, is that the analysis of the partial correlations within the patient groups, in fact, resembled the results that Saboonchi and Lundh (1997) obtained in the non-clinical sample. That is, the correlations between social anxiety and public self-consciousness became non-signi®cant among both social phobics and panic patients when dimensions of perfectionism were controlled for. In addition, some of the correlations between social anxiety and CM and DA in the two patient groups remained signi®cant when public self-consciousness was controlled for. For social phobics, the correlation between social anxiety and doubts about action remained clearly signi®cant and among the panic patients, the correlation between concern over mistakes and social anxiety remained signi®cant. How can these two ®ndings be reconciled? How can public self-consciousness dierentiate better than perfectionism between social phobia and panic disorder, at the same time that the degree of social anxiety is more related to perfectionism than public self-consciousness within each of these patient groups? One possible explanation is that we are dealing with a qualitatively dierent form of social anxiety in social phobia and that public self-consciousness is a dierentiating characteristic of this more severe form of social anxiety. Both `kinds' of social anxiety, however, within their respective ranges, vary in degree as a function of degree of perfectionistic concerns in a way which is relatively independent of the degree of public selfconsciousness. One further result which is worth some attention is that both patient groups scored high on parental criticism, with no signi®cant dierence between the two patient groups. This indicates that parental criticism is a potentially interesting dimension of perfectionism, that should be explored further in its relationships with various kinds of psychopathology. Consistent with the results of other studies (e.g. Juster et al., 1996; Frost & Steketee, 1997), however, personal standards, parental expectation and organization showed to be of less relevance for these anxiety disorders. Finally, it should be noted that the Frost et al. (1990) multidimensional perfectionism scale, which was used in the present study, is one of two equally well-validated and much used instruments. Would the same results have been found if the Hewitt and Flett (1991b) Multidimensional Perfectionism Scale had been used instead? In the present study, all dierences on perfectionism between social phobics and panic patients disappeared when public
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
807
self-consciousnes was controlled for, but would this have been the case also with Hewitt and Flett's scale? The fact that the dierences on concern over mistakes and doubts about action disappeared when public self-consciousness was controlled for suggests that it is only those components of CM and DA which are related to focusing on and monitoring one's external features, as they appear to others, which make up for the dierences on these scales between these two types of anxiety disorder. In other words, social phobics have a higher degree of perfectionism as long as their performance or behaviour is perceived as potentially accessible to, and observed or evaluated by, others. On concern over mistakes, one such component could be the threat of losing others' respect or liking by committing even minor mistakes and on doubts about action it could be a belief that others hold an assumed standard which one's behaviour is unable to meet. These two subscales, however, also contain other items that may be less relevant to social phobia. Maybe a trait similar to what Hewitt and Flett (1991b) call `socially prescribed perfectionism' captures more of the dierence in perfectionism between these two groups of patients. If so, would the expected dierence between these two patient groups on socially prescribed perfectionism also disappear when the eects of public selfconsciousness was controlled? That is, would public self-consciousness show up as a more dierentiating characteristic of social phobia than even socially prescribed perfectionism? Or is it possible that the higher degree of public self-consciousness in social phobics is motivated largely by perceived perfectionistic demands from the social environment? The question of the relationship between social anxiety, public self-consciousness and perfectionism needs to be addressed also with the help of other instruments before any de®nite conclusions are drawn.
References American Psychiatric Association (1987). Diagnostic and statistical manual for mental disorders (3rd ed. rev.). Washington, DC: Author. American Psychiatric Association (1994). Diagnostic and statistical manual for mental disorders (4th ed.). Washington, DC: Author. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561±571. Blatt, S. J., Quinlan, D. M., Pilkonis, P. A., & Shea, M. T. (1995). Impact of perfectionism and need for approval on the brief treatment of depression. Journal of Consulting and Clinical Psychology, 63, 125±132. Brown, T. A., DiNardo, P. A., & Barlow, D. H. (1994). Anxiety Disorder Interview Schedule for DSM IV. University of Albany, Albany, New York: Center for Stress and Anxiety Disorder. Bruch, M. A., & Heimberg, R. G. (1994). Dierences in perception of parental and personal characteristics between generalised and non-generalised social phobics. Journal of Anxiety Disorders, 8, 155±168. Bruch, M. A., Heimberg, R. G., Berger, P., & Collins, T. M. (1989). Social phobia and perception of early parental and personal characteristics. Anxiety Research, 2, 57±65. Clark, D. M. & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg, M. R. Leibowitz, D. A. Hope, & F. R. Schneier (Eds.), Social phobia: diagnosis, assesment and treatment (pp. 63±93). New York: The Guildford Press. DiNardo, P. A. & Barlow, D. H. (1988). The Anxiety Disorder Interview Schedule-Revised (ADIS-R). Albany, NY: Graywind. Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private self-consciousness: assessment and theory. Journal of Consulting and Clinical Psychology, 43, 522±527. Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14, 449±468. Frost, R. O., & Steketee, G. (1997). Perfectionism in obsessive±compulsive disorder patients. Behaviour Research and Therapy, 35, 291± 296. Hewitt, P. L., Flett, G. L., & Turnbull, W. (1994). Borderline personality disorder: an investigation with the multidimensional perfectionism scale. European Journal of Personality Assessment, 10, 28±33.
808
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
Hewitt, P. L., & Flett, G. L. (1991a). Dimensions of perfectionism in unipolar depression. Journal of Abnormal Psychology, 100, 98± 101. Hewitt, P. L., & Flett, G. L. (1991b). Perfectionism in the self and social contexts: conceptualization, assessment, and association with psychopathology. Journal of Personality and Social Psychology, 60, 456±470. Hewitt, P. L., & Flett, G. L. (1993). Dimensions of perfectionism, daily stress, and depression: a test of the speci®c vulnerability hypothesis. Journal of Abnormal Psychology, 100, 98±101. Hope, D. A., & Heimberg, R. G. (1988). Public and pricate self-consciousness and social phobia. Journal of Personality Assesment, 52, 626±639. Joiner, E., Heatherton, T. F., Rudd, M. D., & Schmidt, N. B. (1997). Perfectionism, perceived weight status, and boulimic symptoms: two studies testing a diathesis-stress model. Journal of Aabnormal Psychology, 106, 145±153. Juster, H. R., Frost, R. O., Heimberg, R. G., Holt, C. S., Mattia, J. I., & Faccenda, K. (1996). Social phobia and perfectionism. Personality and Individual Dierences, 21, 403±410. Lundh, L. G., & OÈst, L. G. (1996). Stroop interference, self-focus, and perfectionism in social phobics. Personality and Individual Dierences, 20, 725±731. Nystedt, L., & Smari, J. (1989). Assesment of the Fenigstien, Scheier and Buss Self-Consciousness Scale: a Swedish translation. Journal of Personality Assesment, 53(2), 342±352. Rapee, M. R., & Heimberg, R. G. (1997). A cognitive-behavioral model of anxiety in social phobia. Behaviour Research and Therapy, 8, 741±756. Saboonchi, F., & Lundh, L. G. (1997). Perfectionism, self-consciousness, and anxiety. Personality and Individual Dierences, 22, 921± 928. Schlenker, B. S., & Leary, M. R. (1982). Social anxiety and self-presentation: a conceptualisation and model. Psychological Bulletin, 92, 641±669. Spielberger, C., Gorsuch, R., & Lushene, R. (1970). Manual for State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologist Press. American Psychiatric Association (1988). Author please supply details.
PERGAMON
Behaviour Research and Therapy 37 (1999) 799±808
www.elsevier.com/locate/brat
Perfectionism and self-consciousness in social phobia and panic disorder with agoraphobia Fredrik Saboonchi *, Lars-Gunnar Lundh, Lars-GoÈran OÈst Department of Psychology, Stockholm University, 106 91 Stockholm, Sweden Accepted 22 September 1998
Abstract Social phobics were compared to patients with panic disorder with agoraphobia and normal controls on perfectionism and self-consciousness. On concern over mistakes and doubts about action, social phobics scored higher than patients with panic disorder. Social phobics also demonstrated a higher level of public self-consciousness than patients with panic disorder and when this dierence was controlled for the signi®cant dierences on perfectionism disappeared. Within each patient group, however, perfectionism was more robustly related to social anxiety than was public self-consciousness, which replicates the ®ndings of Saboonchi and Lundh [Saboonchi, F. & Lundh, L. G. (1997). Perfectionism, self-consciousness and anxiety. Personality and Individual Dierences, 22, 921±928.] from a non-clinical sample. The results are discussed in terms of public self-consciousness being a dierentiating characteristic of the more severe kind of social anxiety which is typical of social phobia. # 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction Research on perfectionism has established a wide array of associations between various aspects of perfectionism and numerous forms of psychopathology, including depression (Hewitt & Flett, 1991a, 1993; Blatt, Quinlan, Pilkonis & Shea, 1995), social phobia (Juster et al., 1996; Lundh & OÈst, 1996), obsessive±compulsive disorder (Frost & Steketee, 1997), eating disorders (e.g. Joiner, Heatherton, Rudd & Schmidt, 1997) and borderline personality disorder (Hewitt, Flett & Turnbull, 1994). There are at present two main multidimensional approaches to the empirical study of perfectionism. In the Frost, Marten, Lahart, and Rosenblate (1990) approach, perfectionism is * Corresponding author. E-mail: [email protected] 0005-7967/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 5 - 7 9 6 7 ( 9 8 ) 0 0 1 8 3 - 1
800
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
de®ned as consisting of dimensions of personal standards, concern over mistakes, doubts about the quality of one's own actions, organisation, and parental expectations and criticism. Of these dimensions, concern over mistakes (CM), doubts about action (DA) and parental criticism (PC) have been found to be more associated with psychopathology. In the Hewitt and Flett (1991a) approach, on the other hand, three dimensions of perfectionism are identi®ed: self-oriented perfectionism, other-oriented perfectionism and socially prescribed perfectionism. The latter two dimensions address the interpersonal aspects of perfectionistic demands and socially prescribed perfectionism has been found to be more likely to be related to psychopathology. It can be asked whether perfectionism is a general characteristic of psychopathology or whether dierent forms of psychopathology dier on perfectionism or on some dimensions of perfectionism. With regard to anxiety disorders, higher levels of concern over mistakes, doubts about action and parental criticism were reported in social phobics compared to controls by Juster et al. (1996) and by Lundh and OÈst (1996). Frost and Steketee (1997) found that both patients with obsessive±compulsive disorder (OCD) and patients with panic disorder with agoraphobia had higher levels of concern over mistakes than the controls, but that they did not dier from one another. On doubts about action, the pattern was dierent; OCD patients scored higher than both panic patients and normal controls, who did not dier from one another. On parental criticism, ®nally, the panic patients scored higher than both the OCD patients and the controls, who did not dier from one another. The main purpose of the present study was to compare patients with social phobia, panic disorder with agoraphobia and normal controls on the Frost et al. (1990) Multidimensional Perfectionism Scale. Since social anxiety by de®nition involves a preoccupation with one's social presentation and doubts about one's ability to convey a desired impression on other persons (Schlenker & Leary, 1982; Clark & Wells, 1995; Rapee & Heimberg, 1997), a certain overlap with perfectionistic concerns is apparent. According to the Rapee and Heimberg (1997, pp. 743±744) conceptualization of social phobia, for example, the social phobic ``formulates a prediction of the performance standard or norm which he/she expects the audience to utilize in the given situation. The discrepancy between the person's perception of the audience's appraisal of his/ her performance (appearance and/or behavior) and the person's perception of the audience's standard for the evaluation of his/her appearance and/or behavior, determines the perceived likelihood of negative evaluation from the audience''. The more perfectionistic standards the individual has, or believes that the audience has, the more likely will a negative evaluation appear, which is what the social phobic fears. With these clear indications of associations between social anxiety and perfectionism, two questions can be raised. First, is perfectionism more speci®cally related to social phobia than other psychopathological forms of anxiety? Second, what aspects of perfectionism account for this relationship? Although in a non-clinical population, a similar issue was attended to earlier by Saboonchi and Lundh (1997). Several dimensions of perfectionism, primarily concern over mistakes (CM), doubts about actions (DA) and socially prescribed perfectionism (SPP), showed strong patterns of relationship with a variety of anxiety measures (social anxiety, agoraphobic fears, and fears of bodily injury, death and illness). Social anxiety, however, correlated stronger than agoraphobic fears with CM and stronger than fears of bodily injury,
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
801
death and illness with CM and socially prescribed perfectionism. The present study aims to test the hypothesis that social phobia involves more of perfectionism than panic disorder with agoraphobia, by comparing two groups of patients with these diagnoses and a control group on the Frost et al. (1990) Multidimensional Perfectionism Scale. Another concept that is relevant in this context is that of public self-consciousness. Public self-consciousness is an aspect of self-awareness de®ned by Fenigstein, Scheier, and Buss (1975) as re¯ecting awareness of the self as a social object. This de®nition addresses the preoccupation with, and the degree of attention directed to, aspects of the self that are accessible to others such as looks, impressions etc. Social anxiety has been found to correlate positively with public self-consciousness (Fenigstein et al., 1975; Hope & Heimberg, 1988) and individuals with social phobia have been found to score higher on public self-consciousness than normal controls and individuals with other anxiety disorders (Bruch, Heimberg, Berger & Collins, 1989; Bruch & Heimberg, 1994). In the earlier study by Saboonchi and Lundh (1997), however, ®ndings indicated a non-signi®cant pattern of relationship between public self-consciousness and social anxiety when dimensions of perfectionism were controlled for. Since these results were found in a non-clinical sample of subjects, it was unclear whether the same results would be obtained with a clinical sample of patients with a diagnosis of social phobia. A second purpose of the present paper, therefore, was to explore this issue. In summary, three hypotheses are tested by this study: (a) both social phobia and panic disorder with agoraphobia are characterized by elevated degrees of perfectionism, particularly on the subscales concern over mistakes, doubts about action and parental criticism. (b) Social phobia is characterized by a higher degree of perfectionism than panic disorder with agoraphobia. (c) Social phobia is also characterized by higher scores on public selfconsciousness than both panic disorder with agoraphobia and normal controls; this dierence, however, disappears when degree of perfectionism is controlled for. On the other hand, the dierences on perfectionism remain when public self-consciousness is controlled for.
2. Method 2.1. Subjects The subjects were 52 social phobics (8 males and 44 females aged 33.6 (S.D. = 8.8) yr), 55 patients with a diagnosis of panic disorder with agoraphobia (19 males and 36 females with a mean age of 34.5 (S.D. = 10.7) yr) and 113 control subjects (27 males and 86 females aged 30.6 (S.D. = 9.8) yr). The social phobics were part of a treatment study with cognitivebehaviour therapy, and as part of that study they were interviewed by trained clinical psychologists with the revised version of the Anxiety Disorders Interview Schedule (ADIS-R; DiNardo & Barlow, 1988); all patients met the DSM-III-R criteria (American Psychiatric Association, 1987) for social phobia. The panic patients were part of another treatment study with cognitive-behaviour therapy and as part of that study they were interviewed by a trained clinical psychologist with the Anxiety Disorders Interview Schedule (ADIS-IV; Brown, DiNardo & Barlow, 1994); all of the patients met the DMS-IV criteria (American Psychiatric Association, 1994) for panic disorder with agoraphobia. Potential Ss in both studies were
802
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
excluded if they received a diagnosis of primary depression, bipolar disorder, psychotic disorder or active drug or alcohol dependence. The control group were volunteers, who were recruited in two ways: (a) with advertisements in a living area adjacent to the university (36 females and 7 males); these subjects were paid for their participation and (b) undergraduate students in psychology (50 females and 20 males), who received course credit for their participation. All patients were administered the questionnaires before they started therapy. 2.2. Psychometric measures 2.2.1. Multidimensional perfectionism scale (MPS) The MPS (Frost et al., 1990) contains 35 items in the form of statements with a Likert type 5-point response format from `strongly disagree' to `strongly agree'. These 35 items are being divided into 6 subscales tapping six dimensions of perfectionism, i.e. concern over mistakes, personal standards, parental expectations, parental criticism, doubts about actions and organisation. 2.2.2. Self-consciousness scale (SCS) This instrument was a Swedish translation (Nystedt & Smari, 1989) of the Fenigstein et al. (1975) Self-Consciousness Scale, which includes a subscale for private self-consciousness (i.e. the disposition to attend to one's own thoughts, feelings, moods and attitudes), a subscale for public self-consciousness (i.e. the disposition to focus on outwardly observable aspects of the self, such as physical appearance and overt behaviour) and a subscale of social anxiety. 2.2.3. State-Trait Anxiety Inventory, Trait Form (STAI-T) The STAI-T (Spielberger, Gorsuch & Lushene, 1970) contains 20 items that assess enduring symptoms of anxiety. 2.2.4. Beck Depression Inventory (BDI) The BDI (Beck, Ward, Mendelson, Mock & Erbaugh, 1961) is a 21-item self-report questionnaire that assesses the presence and severity of cognitive, behavioural, aective and somatic symptoms of depression. Each item consists of a 0 to 3 self-rating, with increasing scores indicating greater symptom severity.
3. Results Means and standard deviations for the two groups of patients and controls are presented in Table 1. As can be seen, there were no scores available on BDI or STAI-T for the control group; accordingly, the analysis involving these measures was only performed on the two patient groups. A MANOVA on all the subscales of the MPS and SCS with all the three groups revealed a highly signi®cant eect of Group (Wilks' Lambda = 0.613, F(18, 416) = 6.41, P < 0.001). As is displayed in Table 1, the follow-up univariate tests revealed signi®cant eects of group on
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
803
Table 1 Means and standard deviations for social phobics, panic patients and control subjects Social phobics
Panic patients
Controls
M
S.D.
M
S.D.
M
S.D.
F-value
100.7 26.7a 10.0a 21.8 9.6a 11.8 20.8
21.6 7.6 3.5 6.0 4.6 5.4 4.5
91.0 22.7b 8.4b 19.6 9.3a 11.4 19.6
24.3 7.8 3.3 6.2 4.8 5.8 5.1
84.4 19.7c 7.5b 20.5 7.0b 10.1 19.5
17.9 6.4 3.2 5 3.6 4.3 5.2
17.43* 10.10* 2.11 9.36* 2.86 2.33
SCS Public self-consciousness Private self-consciousness Social anxiety
22.8a 25.8 18.1a
3.9 6.4 4.4
18.2b 23.9 12.3b
5.2 5.4 5.7
17.8b 23.6 9.8c
5.1 6.7 5.5
19.80* 4.57 43.41*
Beck Depression Inventory State-Trait Anxiety Inventory-Trait
16.8 53.3
9.2 8.7
13.4 49.7
8.3 10.2
MPS Total Concern over mistakes Doubts about actions Personal standards Parental criticism Parental expectation Organization
± ±
ANOVA
± ±
3.79 3.83
*
P < 0.001.Means with dierent superscripts are signi®cantly dierent at the 0.05 level.
social anxiety, public self-consciousness, concern over mistakes, doubts about action and parental criticism. To examine these dierences further, one-way ANOVAS were performed on these MPS and SCS subscales and multiple comparisons among the three scores were done using Schee Ftest. The results showed that the social phobics scored signi®cantly higher than the controls on social anxiety, public self-consciousness, concern over mistakes, doubts about action and parental criticism. In addition, the social phobics scored signi®cantly higher than the panic patients on social anxiety, public self-consciousness, concern over mistakes and doubts about action. The panic patients, in turn, scored signi®cantly higher than the controls on parental criticism, social anxiety and concern over mistakes. The two patient groups were compared on the Beck Depression Inventory and the StateTrait Anxiety Inventory-Trait version by means of a two-tailed t-test. The social phobics had a non-signi®cant tendency to higher scores on both the BDI, t(104) = 1.95, P = 0.054, and the STAI-T, t(104) = 1.96, P = 0.053. In order to test the hypothesis that the group dierences on perfectionism between the two patient groups would remain when public self-consciousness was controlled for, a multivariate analysis of covariance on concern over mistakes and doubts about action was carried out (since these were the only dimensions of perfectionism on which social phobics scored higher than panic patients), with public self-consciousness included as a covariate. The multivariate test and both univariate tests were non-signi®cant, i.e. no signi®cant dierences remained, neither on concern over mistakes, F(1, 103) = 0.08, nor on doubts about action, F(1, 103) = 0.17. However, when analysis of covariance was performed on public self-consciousness
804
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
Table 2 Correlations and partial correlations (controlled for public self-consciousness) between aspects of perfectionism and measures of anxiety and depression Social phobics
SCS social anxiety STAI-T BDI
Panic patients
Controls
CM
DA
CM
DA
CM
DA
0.32* (0.23) 0.33* (0.28*) 0.30* (0.28*)
0.44** (0.36*) 0.53** (0.50**) 0.18 (0.15)
0.46* (0.21) 0.23 (0.26) 0.27 (0.12)
0.34** (0.21) 0.29* (0.26) 0.21 (0.12)
0.48** (0.43**) ± ±
0.54** (0.50**) ± ±
Values in parentheses are partial correlations controlling for SCS public self-consciousness. CM = concern over mistakes, DA = doubts about action.*P < 0.05.**P < 0.01.
with concern over mistakes and doubts about action as covariates, the group eect remained highly signi®cant, F(1, 102) = 16.24, P < 0.001. Regarding the comparisons with the controls, however, when a similar MANOVA and its follow up univariate tests on the subscales of MPS were performed with public selfconsciousness included as covariate, the dierences that were found in the initial analysis between each patient group and the controls remained signi®cant. Correlations and partial correlations (controlling for public self-consciousness) between CM, DA and measures of anxiety (social anxiety and STAIT-T) and depression (BDI) for the two patient groups and the controls (only social anxiety) are presented in Table 2. For social phobics CM correlated signi®cantly with both measures of anxiety and with depression (P < 0.05). With the exception for social anxiety these correlations remained signi®cant when public self-consciousness was controlled for. DA correlated signi®cantly with social anxiety and STAIT-T and these correlations remained signi®cant (P < 0.01) when controlled for public self-consciousness. The panic patients demonstrated a somewhat dierent pattern. CM correlated signi®cantly (P < 0.01) with SCS social anxiety and this correlation remained when controlled for public self-consciousness. DA correlated signi®cantly (P < 0.05) with SCS social anxiety and STAITT. These correlations, however, were rendered non-signi®cant when public self-consciousness was controlled for. Concerning the controls, both CM and DA correlated signi®cantly with SCS social anxiety (P < 0.01) and these correlations still remained signi®cant (P < 0.01) when public selfconsciousness was controlled for. Correlations and partial correlations (controlling for concern over mistakes and doubts about action) between public self-consciousness and measures of anxiety and depression are displayed in Table 3. For social phobics public self-consciousness correlated signi®cantly with social anxiety; when CM and DA were controlled for, however, this correlation became non-signi®cant. For panic patients there were signi®cant correlations between public self-consciousness and social anxiety (P < 0.01) and BDI (P < 0.05). The latter, however, disappeared when CM and DA was controlled for. For the controls public self-consciousness correlated signi®cantly with SCS
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
805
Table 3 Correlations and partial correlations (controlled for concern over mistakes and doubts about action) between public self-consciousness and measures of anxiety and depression Public self-consciousness
SCS social anxiety State-Trait Anxiety Inventory-Trait Beck Depression Inventory
panic patients
social phobics
controls
0.39** (0.18) 0.18 (0.05) 0.29* (0.19)
0.33** (0.16) 0.20 (ÿ0.03) 0.11 (ÿ0.02)
0.28** (0.11) ± ±
Values in parentheses are partial correlations controlling for concern over mistakes and doubt about actions subscales of MPS.*P < 0.05.**P < 0.01.
social anxiety ( p < 0.01). This correlation, however, was no longer signi®cant when CM and DA were controlled for.
4. Discussion The results of the present study showed that the social phobics scored higher than the normal controls on three dimensions of perfectionism, i.e. concern over mistakes, doubts about action and parental criticism. They also scored higher than patients with a diagnosis of panic disorder with agoraphobia on two of these dimensions, i.e. concern over mistakes and doubts about action. The panic patients, however, also showed evidence of an elevated degree of perfectionism, since they scored higher than the controls on concern over mistakes and parental criticism. These results are consistent with earlier studies, both with regard to social phobics (e.g. Juster et al., 1996) and panic patients (Frost & Steketee, 1997). The results also supported the hypothesis that social phobia involves more of perfectionism than panic disorder, at least with regard to the dimensions concern over mistakes and doubts about action. In addition, the social phobics scored higher than both the controls and the panic patients on social anxiety and on public self-consciousness. It is interesting to note that, although the panic patients scored higher than the controls on social anxiety, they did not dier from the controls on the public self-consciousness scale. This indicates that the degree of public selfconsciousness, i.e. the individual's tendency to focus on how he/she appears to others, is a highly dierentiating characteristic between social phobics and panic patients. The two patient groups, in fact, diered considerably more on public self-consciousness than on the dimensions of perfectionism and the hypothesis that the group dierences between the two patient groups on perfectionism would remain when public self-consciousness was controlled for was ¯atly contradicted. Quite to the contrary, all the group dierences on perfectionism disappeared when the eects of public self-consciousness were controlled. Moreover, the social phobics' higher scores on public self-consciousness, as compared with the panic patients, remained when perfectionism was controlled for. This suggests that the social phobics' stronger tendency to focus on how they appear to others, as compared with the
806
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
panic patients', is not reducible to their tendency to doubt the quality of their behaviour and the harshness that characterizes their perfectionistic concern over mistakes. The hypothesis that this dierence is determined by the perfectionistic aspects of this trait got no support whatever. On the contrary, the results suggest that social phobia is essentially characterized by a higher level of public self-consciousness than panic disorder with agoraphobia, in a way which is not motivated by perfectionistic concerns. This result, at ®rst sight, seems to be totally at odds with the results in the study by Saboonchi and Lundh (1997), where the correlation between social anxiety and perfectionism remained when public self-consciousness was controlled for, whereas the correlation between social anxiety and public self-consciousness was no longer signi®cant when perfectionism was controlled for. One dierence between that study and the present one is that Saboonchi and Lundh (1997) used a non-clinical sample. This might suggest that the relationships between social anxiety, perfectionism and public self-consciousness are dierent in clinical and nonclinical samples. What speaks against this conclusion, however, is that the analysis of the partial correlations within the patient groups, in fact, resembled the results that Saboonchi and Lundh (1997) obtained in the non-clinical sample. That is, the correlations between social anxiety and public self-consciousness became non-signi®cant among both social phobics and panic patients when dimensions of perfectionism were controlled for. In addition, some of the correlations between social anxiety and CM and DA in the two patient groups remained signi®cant when public self-consciousness was controlled for. For social phobics, the correlation between social anxiety and doubts about action remained clearly signi®cant and among the panic patients, the correlation between concern over mistakes and social anxiety remained signi®cant. How can these two ®ndings be reconciled? How can public self-consciousness dierentiate better than perfectionism between social phobia and panic disorder, at the same time that the degree of social anxiety is more related to perfectionism than public self-consciousness within each of these patient groups? One possible explanation is that we are dealing with a qualitatively dierent form of social anxiety in social phobia and that public self-consciousness is a dierentiating characteristic of this more severe form of social anxiety. Both `kinds' of social anxiety, however, within their respective ranges, vary in degree as a function of degree of perfectionistic concerns in a way which is relatively independent of the degree of public selfconsciousness. One further result which is worth some attention is that both patient groups scored high on parental criticism, with no signi®cant dierence between the two patient groups. This indicates that parental criticism is a potentially interesting dimension of perfectionism, that should be explored further in its relationships with various kinds of psychopathology. Consistent with the results of other studies (e.g. Juster et al., 1996; Frost & Steketee, 1997), however, personal standards, parental expectation and organization showed to be of less relevance for these anxiety disorders. Finally, it should be noted that the Frost et al. (1990) multidimensional perfectionism scale, which was used in the present study, is one of two equally well-validated and much used instruments. Would the same results have been found if the Hewitt and Flett (1991b) Multidimensional Perfectionism Scale had been used instead? In the present study, all dierences on perfectionism between social phobics and panic patients disappeared when public
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
807
self-consciousnes was controlled for, but would this have been the case also with Hewitt and Flett's scale? The fact that the dierences on concern over mistakes and doubts about action disappeared when public self-consciousness was controlled for suggests that it is only those components of CM and DA which are related to focusing on and monitoring one's external features, as they appear to others, which make up for the dierences on these scales between these two types of anxiety disorder. In other words, social phobics have a higher degree of perfectionism as long as their performance or behaviour is perceived as potentially accessible to, and observed or evaluated by, others. On concern over mistakes, one such component could be the threat of losing others' respect or liking by committing even minor mistakes and on doubts about action it could be a belief that others hold an assumed standard which one's behaviour is unable to meet. These two subscales, however, also contain other items that may be less relevant to social phobia. Maybe a trait similar to what Hewitt and Flett (1991b) call `socially prescribed perfectionism' captures more of the dierence in perfectionism between these two groups of patients. If so, would the expected dierence between these two patient groups on socially prescribed perfectionism also disappear when the eects of public selfconsciousness was controlled? That is, would public self-consciousness show up as a more dierentiating characteristic of social phobia than even socially prescribed perfectionism? Or is it possible that the higher degree of public self-consciousness in social phobics is motivated largely by perceived perfectionistic demands from the social environment? The question of the relationship between social anxiety, public self-consciousness and perfectionism needs to be addressed also with the help of other instruments before any de®nite conclusions are drawn.
References American Psychiatric Association (1987). Diagnostic and statistical manual for mental disorders (3rd ed. rev.). Washington, DC: Author. American Psychiatric Association (1994). Diagnostic and statistical manual for mental disorders (4th ed.). Washington, DC: Author. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561±571. Blatt, S. J., Quinlan, D. M., Pilkonis, P. A., & Shea, M. T. (1995). Impact of perfectionism and need for approval on the brief treatment of depression. Journal of Consulting and Clinical Psychology, 63, 125±132. Brown, T. A., DiNardo, P. A., & Barlow, D. H. (1994). Anxiety Disorder Interview Schedule for DSM IV. University of Albany, Albany, New York: Center for Stress and Anxiety Disorder. Bruch, M. A., & Heimberg, R. G. (1994). Dierences in perception of parental and personal characteristics between generalised and non-generalised social phobics. Journal of Anxiety Disorders, 8, 155±168. Bruch, M. A., Heimberg, R. G., Berger, P., & Collins, T. M. (1989). Social phobia and perception of early parental and personal characteristics. Anxiety Research, 2, 57±65. Clark, D. M. & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg, M. R. Leibowitz, D. A. Hope, & F. R. Schneier (Eds.), Social phobia: diagnosis, assesment and treatment (pp. 63±93). New York: The Guildford Press. DiNardo, P. A. & Barlow, D. H. (1988). The Anxiety Disorder Interview Schedule-Revised (ADIS-R). Albany, NY: Graywind. Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private self-consciousness: assessment and theory. Journal of Consulting and Clinical Psychology, 43, 522±527. Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14, 449±468. Frost, R. O., & Steketee, G. (1997). Perfectionism in obsessive±compulsive disorder patients. Behaviour Research and Therapy, 35, 291± 296. Hewitt, P. L., Flett, G. L., & Turnbull, W. (1994). Borderline personality disorder: an investigation with the multidimensional perfectionism scale. European Journal of Personality Assessment, 10, 28±33.
808
F. Saboonchi et al. / Behaviour Research and Therapy 37 (1999) 799±808
Hewitt, P. L., & Flett, G. L. (1991a). Dimensions of perfectionism in unipolar depression. Journal of Abnormal Psychology, 100, 98± 101. Hewitt, P. L., & Flett, G. L. (1991b). Perfectionism in the self and social contexts: conceptualization, assessment, and association with psychopathology. Journal of Personality and Social Psychology, 60, 456±470. Hewitt, P. L., & Flett, G. L. (1993). Dimensions of perfectionism, daily stress, and depression: a test of the speci®c vulnerability hypothesis. Journal of Abnormal Psychology, 100, 98±101. Hope, D. A., & Heimberg, R. G. (1988). Public and pricate self-consciousness and social phobia. Journal of Personality Assesment, 52, 626±639. Joiner, E., Heatherton, T. F., Rudd, M. D., & Schmidt, N. B. (1997). Perfectionism, perceived weight status, and boulimic symptoms: two studies testing a diathesis-stress model. Journal of Aabnormal Psychology, 106, 145±153. Juster, H. R., Frost, R. O., Heimberg, R. G., Holt, C. S., Mattia, J. I., & Faccenda, K. (1996). Social phobia and perfectionism. Personality and Individual Dierences, 21, 403±410. Lundh, L. G., & OÈst, L. G. (1996). Stroop interference, self-focus, and perfectionism in social phobics. Personality and Individual Dierences, 20, 725±731. Nystedt, L., & Smari, J. (1989). Assesment of the Fenigstien, Scheier and Buss Self-Consciousness Scale: a Swedish translation. Journal of Personality Assesment, 53(2), 342±352. Rapee, M. R., & Heimberg, R. G. (1997). A cognitive-behavioral model of anxiety in social phobia. Behaviour Research and Therapy, 8, 741±756. Saboonchi, F., & Lundh, L. G. (1997). Perfectionism, self-consciousness, and anxiety. Personality and Individual Dierences, 22, 921± 928. Schlenker, B. S., & Leary, M. R. (1982). Social anxiety and self-presentation: a conceptualisation and model. Psychological Bulletin, 92, 641±669. Spielberger, C., Gorsuch, R., & Lushene, R. (1970). Manual for State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologist Press. American Psychiatric Association (1988). Author please supply details.